1
Allen, M. P. (2004). Introduction to Molecular Dynamics Simulation. 4-4. Wu, S., & Tran, L. (2017). Molecular Dynamics Simulation of Epoxy Resin System. 2-2. S. Plimpton, Fast Parallel Algorithms for Short-Range Molecular Dynamics, J Comp Phys, 117, 1-19 (1995) http://lammps.sandia.gov L. Martínez, R. Andrade, E. G. Birgin, J. M. Martínez. Packmol: A package for building initial configurations for molecular dynamics simulations. Journal of Computational Chemistry, 30(13):2157-2164, 2009. M. Valiev, E.J. Bylaska, N. Govind, K. Kowalski, T.P. Straatsma, H.J.J. van Dam, D. Wang, J. Nieplocha, E. Apra, T.L. Windus,W.A. de Jong, "NWChem: a comprehensive and scalable open-source solution for large scale molecular simulations" Comput. Phys. Commun. 181, 1477 (2010) Humphrey, W., Dalke, A. and Schulten, K., "VMD - Visual Molecular Dynamics", J. Molec. Graphics, 1996, vol. 14, pp. 33-38. Molecular Dynamics Simulations of Epoxy Resin Systems to Study Physical Properties Chase Brooks - New Mexico State University Hongwei Zuo - City University of Hong Kong References Introduction The epoxy resin systems were modeled by the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) package, NWChem, Visual Molecular Dynamics (VMD), and Packmol. Here are the steps in detail. Creating molecules individually and supplementing with topological information in VMD; Structural optimization (Figure 4) and analysis of partial charge distribution using quantum mechanics simulation in NWChem; Parameterization of atom types, bond (stretch), angle (bend), dihedral (torsion) and improper dihedral types under the CHARMM36 Force Field; Packing of molecules into initial simulation box (Figure 5) using Packmol and energy minimization in LAMMPS; ● Reacted group activation (Figure 6) and partial charge rearrangement; Equilibration of molecules under NVE, NVT, and NPT ensembles; Crosslinking formation (Figure 7) based on cut-off distance using the LAMMPS fix bond/create command The epoxy resin systems at various crosslinking percentages were created by the repetition of energy minimization and bond formation with different time steps under LAMMPS package. Periodic boundary conditions were applied to save computational cost while satisfying the requirements of bulk phase. Preparation of themoset epoxy resin systems Thus far we have been able to successfully construct the six molecules needed for our model epoxy resin systems, optimized their geometries, and created several systems under similar conditions but at various stages of crosslinking. We have begun measuring several physical properties of our systems, and hope to determine how these properties vary for different crosslinking percentages. To further our research we would like to study methods of making our model systems more realistic. One strategy we would like to employ to make our simulation more accurate is to increase the size of our system and crosslink a greater number of molecules. Our belief is that by making the system larger the quantities we are measuring at the microscopic level will more closely resemble a macroscopic, real world application. Conclusions and future steps Molecular dynamics (MD) is a computer simulation method for studying the physical movement of atoms and molecules. This project focuses on modeling thermoset epoxy resin systems, which are widely used in engineering applications, and studying their physical properties using molecular dynamics simulations. Epoxy resins are molecular systems that include an epoxide- group containing base resin as well as a an amine bonding curative, with additional modifiers sometimes included for other properties. PRO-SET M1002 resin and M2046 hardener (Figure 1) were used as raw materials to prepare the systems. They were mixed at a high temperature and were crosslinked (Figure 2) in what is known as a curing process to form epoxy resin systems at various crosslinking degrees under different time steps. We applied the CHARMM36 force field (Figure 3) to the MD simulation of the reaction and future measurement, where the parameterization of these coefficients affects how realistic our systems are. There are limitations to the accuracy of our model when compared with real experiment, but it saves experimental resources and time. This helps us to study the relationships among physical properties and chemical structures by an alternative method. Figure 3. Parameterization – stretch, bend, torsion Figure 4. Structural optimization Optimized Original Bonds Dihedrals Angles Figure 6. Epoxide and amine group activation Figure 5. Simulation box Figure 7. Sample crosslinking C-N bond Measuring physical properties We have prepared multiple epoxy resin systems at various crosslinking percentages. Our goal now is to run simultaneous simulations on each of these systems to measure physical properties that are significant to engineering applications and determine to what degree these quantities vary based on crosslinking percentage. The physical properties we want to measure include: ● Thermal expansion coefficient—a measure of how the volume of the system changes in response to a change in temperature ● Glass transition temperature—the temperature at which the system changes from a hard, “glassy” state to a viscous, rubbery state. ● Isothermal compressibility—how volume changes with respect to a change in the applied external pressure ● Thermal conductivity—a measure of how well heat flows through the system Thermodynamic quantities (e.g. temperature, pressure, volume) of our system are measured by calculating ensemble averages in equilibrium. For the above physical properties, we need to apply different ensembles with different quantities fixed by external constraints. The LAMMPS package allows us to apply external fixes on our system so we can study the canonical, microcanonical, and isothermal-isobaric ensembles—NVT, NVE, and NPT respectively. Figure 1. M1002-M2046 pro-set epoxy resin system Acknowldegements This project was sponsored by the National Science Foundation through Research Experience for Undergraduates (REU) award, with additional support from the Joint Institute of Computational Sciences at University of Tennessee Knoxville. This project used allocations from the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by the National Science Foundation. In addition, the computing work was also performed on technical workstations donated by the BP High Performance Computing Team. Figure 2. Curing reaction Figure 3. CHARMM36 force field potential Mentors: Dr. Lonnie Crosby Dr. Kwai Wong

Molecular Dynamics Simulations of Epoxy Resin Systems to

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Molecular Dynamics Simulations of Epoxy Resin Systems to

● Allen, M. P. (2004). Introduction to Molecular Dynamics Simulation. 4-4.● Wu, S., & Tran, L. (2017). Molecular Dynamics Simulation of Epoxy Resin System. 2-2.● S. Plimpton, Fast Parallel Algorithms for Short-Range Molecular Dynamics, J Comp Phys, 117, 1-19 (1995) http://lammps.sandia.gov● L. Martínez, R. Andrade, E. G. Birgin, J. M. Martínez. Packmol: A package for building initial configurations for molecular dynamics simulations.

Journal of Computational Chemistry, 30(13):2157-2164, 2009.● M. Valiev, E.J. Bylaska, N. Govind, K. Kowalski, T.P. Straatsma, H.J.J. van Dam, D. Wang, J. Nieplocha, E. Apra, T.L. Windus, W.A. de Jong,

"NWChem: a comprehensive and scalable open-source solution for large scale molecular simulations" Comput. Phys. Commun. 181, 1477 (2010)● Humphrey, W., Dalke, A. and Schulten, K., "VMD - Visual Molecular Dynamics", J. Molec. Graphics, 1996, vol. 14, pp. 33-38.

MolecularDynamicsSimulationsofEpoxyResinSystemstoStudyPhysicalProperties

ChaseBrooks- NewMexicoStateUniversityHongweiZuo- CityUniversityofHongKong

References

IntroductionThe epoxy resin systems were modeled by the Large-scale Atomic/MolecularMassively Parallel Simulator (LAMMPS) package, NWChem, Visual MolecularDynamics (VMD), and Packmol. Here are the steps in detail.

● Creating molecules individually and supplementing with topologicalinformation in VMD;

● Structural optimization (Figure 4) and analysis of partial charge distributionusing quantum mechanics simulation in NWChem;

● Parameterization of atom types, bond (stretch), angle (bend), dihedral(torsion) and improper dihedral types under the CHARMM36 Force Field;

● Packing of molecules into initial simulation box (Figure 5) using Packmoland energy minimization in LAMMPS;

● Reacted group activation (Figure 6) and partial charge rearrangement;● Equilibration of molecules under NVE, NVT, and NPT ensembles;● Crosslinking formation (Figure 7) based on cut-off distance using the

LAMMPS fix bond/create command

The epoxy resin systems at various crosslinking percentages were created bythe repetition of energy minimization and bond formation with different timesteps under LAMMPS package. Periodic boundary conditions were applied tosave computational cost while satisfying the requirements of bulk phase.

Preparationofthemosetepoxyresinsystems

Thus far we have been able to successfully construct the six moleculesneeded for our model epoxy resin systems, optimized their geometries, andcreated several systems under similar conditions but at various stages ofcrosslinking. We have begun measuring several physical properties of oursystems, and hope to determine how these properties vary for differentcrosslinking percentages.To further our research we would like to study methods of making our modelsystems more realistic. One strategy we would like to employ to make oursimulation more accurate is to increase the size of our system and crosslink agreater number of molecules. Our belief is that by making the system largerthe quantities we are measuring at the microscopic level will more closelyresemble a macroscopic, real world application.

Conclusionsandfuturesteps

Molecular dynamics (MD) is a computer simulation method for studying thephysical movement of atoms and molecules. This project focuses on modelingthermoset epoxy resin systems, which are widely used in engineeringapplications, and studying their physical properties using molecular dynamicssimulations. Epoxy resins are molecular systems that include an epoxide-group containing base resin as well as a an amine bonding curative, withadditional modifiers sometimes included for other properties. PRO-SETM1002 resin and M2046 hardener (Figure 1) were used as raw materials toprepare the systems. They were mixed at a high temperature and werecrosslinked (Figure 2) in what is known as a curing process to form epoxy resinsystems at various crosslinking degrees under different time steps. We appliedthe CHARMM36 force field (Figure 3) to the MD simulation of the reactionand future measurement, where the parameterization of these coefficientsaffects how realistic our systems are. There are limitations to the accuracy ofour model when compared with real experiment, but it saves experimentalresources and time. This helps us to study the relationships among physicalproperties and chemical structures by an alternative method.

Figure3. Parameterization– stretch,bend,torsion

Figure4. Structuraloptimization

Optimized

Original

Bonds DihedralsAngles

Figure6. Epoxideandaminegroupactivation

Figure5. Simulationbox Figure7. SamplecrosslinkingC-Nbond

MeasuringphysicalpropertiesWe have prepared multiple epoxy resin systems at various crosslinkingpercentages. Our goal now is to run simultaneous simulations on each ofthese systems to measure physical properties that are significant toengineering applications and determine to what degree these quantities varybased on crosslinking percentage. The physical properties we want tomeasure include:● Thermal expansion coefficient—a measure of how the volume of the

system changes in response to a change in temperature● Glass transition temperature—the temperature at which the system

changes from a hard, “glassy” state to a viscous, rubbery state.● Isothermal compressibility—how volume changes with respect to a change

in the applied external pressure● Thermal conductivity—a measure of how well heat flows through the

systemThermodynamic quantities (e.g. temperature, pressure, volume) of oursystem are measured by calculating ensemble averages in equilibrium. Forthe above physical properties, we need to apply different ensembles withdifferent quantities fixed by external constraints. The LAMMPS package allowsus to apply external fixes on our system so we can study the canonical,microcanonical, and isothermal-isobaric ensembles—NVT, NVE, and NPTrespectively.

Figure 1. M1002-M2046 pro-set epoxy resin system

AcknowldegementsThis project was sponsored by the National Science Foundation throughResearch Experience for Undergraduates (REU) award, with additionalsupport from the Joint Institute of Computational Sciences at University ofTennessee Knoxville. This project used allocations from the Extreme Scienceand Engineering Discovery Environment (XSEDE), which is supported by theNational Science Foundation. In addition, the computing work was alsoperformed on technical workstations donated by the BP High PerformanceComputing Team.

Figure 2. Curing reaction Figure 3. CHARMM36 force field potential

Mentors:Dr.LonnieCrosbyDr.Kwai Wong